The long-term objective is to understand the genetic origins and biological functions of unique, (i.e. individually distinct) antigens. These antigens lead to rejection of regressor tumors when transplanted into naive mice and of progressor tumors when transplanted into pre- immunized mice. Two types of unique antigens recognized by CD8+ T cells will be analyzed: those that must be lost by regressors before they can grow in a naive mice and those that can be retained by progressor variants. In the first Aim, it will be determined whether expression of a somatically mutated p68 peptide indeed gives rise to the rejection of the regressor tumor, by analyzing whether re-expression of the mutant gene by an antigen-loss variant reverts from a progressor to a regressor phenotype. If this is observed, it will further be determined whether re- expression of this antigen is sufficient even in established cancer to elicit tumor rejection. In Aim 2, it will be studies which changes in p68 helicase are observed in various cancers (in particular mutations and up- regulation of expression), and whether the already observed point mutation in the IQ domain of the gene results in oncogenic properties, while the wild-type p68 acts as a tumor suppressor gene. The main goal of Aim 3 is to confirm or refute the general principal that unique rejection antigens of UV-induced tumors are due to somatic mutations in a protein expressed at high levels in the cancer cells. The CD8+ T cell-defined unique 5117-RE tumor antigen which is consistently lost by progressor variants will be used as the model. If a mutant gene is identified, the potential significance of the mutations for the malignant process will be studied. The fourth Aim will be to identify the genetic origins of unique antigens that are retained on primary and variant progressor tumors and are insufficient to elicit tumor rejection by naive mice. It will be examined whether these antigens are also encoded by somatic tumor-specific mutations and whether these mutant peptides are equally or less immunogenic than mutant peptides expressed only by regressor tumors. Finally, we will determine whether loss of heterozygosity of the mutant genes encoding unique antigens on the progressor tumors is responsible for the extreme resistance of certain CTL epitopes to immunoselection, and we will analyze in standard assays the relevance of the mutant genes to the development of malignancy.